基于Doc2Vec和LDA模型融合文献质量的学术论文推荐研究

• 1:中国矿业大学图书馆

基于Doc2Vec和LDA模型融合文献质量的学术论文推荐研究.pdf

摘要(Abstract)：

为解决海量的电子资源给读者带来“信息过载”的困扰，采用基于内容的推荐算法为读者推荐内容适配、质量优良的学术论文.考虑论文文本的上下文语义、词序及全局主题信息，首先采用Doc2Vec和LDA(Latent Dirichlet Allocation)混合语义模型训练候选论文集摘要语料库，学习得到每篇论文的文本向量，其次利用K-Means算法对候选论文集进行聚类，然后探寻目标论文所属簇的类群成员作为待推荐论文，最后融合文献质量权重进行相似度计算并排序，从而得到TOP-N近邻推荐结果.以CNKI图书情报类期刊论文作为语料库，通过实证分析，采用的混合模型与传统的TF-IDF(Term Frequency-Inverse Document Frequency)、Word2Vec、LDA 3种模型相比，推荐结果的精确率较高、排序差异度低，达到良好的推荐效果.
The content-based recommendation algorithm is used to recommend academic papers with adaptive content and high quality for readers, so as to solve the problems of information overload caused by massive electronic resources. The context, word order & global topic information of the thesis text are taken into consideration. Firstly, Doc2Vec and LDA hybrid semantic model are used to train the summary corpus of candidate thesis sets, and the text vector of each thesis is learned. Then, the candidate thesis sets are clustered by K-means algorithm, and then the cluster members of the target papers are searched as the papers to be recommended, Finally, the similarity is calculated and sorted by fusing the literature quality weight, so as to obtain the TOP-N nearest neighbor recommendation results. Taking CNKI library & information journal pap as the corpus, an empirical analysis is conducted. Word2Vec & LDA models, the hybrid model adopted in this paper, compared with the traditional TF-IDF, the hybrid model adopted in this paper has higher accuracy and lower ranking difference, and achieves good recommendation results.

关键词(KeyWords)：学术论文;混合语义模型;文献质量;推荐
academic papers;mixed semantic model;literature quality;recommend

基金项目(Foundation):江苏省高校哲学社会科学研究项目(2022SJYB1129);; 国家社科基金(22BTQ023)

作者(Authors):王大阜;邓志文;贾志勇;王静;
Wang Dafu;Deng Zhiwen;Jia Zhiyong;Wang Jing;Library, China University of Mining and Technology;

DOI:10.16366/j.cnki.1000-2367.2023.04.005

参考文献(References)：

• [1] 蒲姗姗，何燕.个性化学术资源推荐研究：现状、特点及展望[J].图书馆学研究，2019(16):9-17.PU S S,HE Y.Personalized academic resource recommendation:research status,features,and prospects[J].Research on Library Science,2019(16):9-17.

• [2] 王仁武，张文慧.学术用户画像的行为与兴趣标签构建与应用[J].现代情报，2019,39(9):54-63.WANG R W,ZHANG W H.Behavior and interest labeling construction and application of academic user portraits[J].Journal of Modern Information,2019,39(9):54-63.

• [3] 刘岩.基于机器学习算法的图书馆书目协同推荐系统[J].现代电子技术，2020,43(14):180-182.LIU Y.Library bibliographic collaborative recommendation system based on machine learning algorithm[J].Modern Electronics Technique,2020,43(14):180-182.

• [4] 李澎林，洪之渊，李伟.基于兴趣度与类型因子的高校图书推荐算法[J].浙江工业大学学报，2019,47(4):425-429.LI P L,HONG Z Y,LI W.Book recommendation algorithm base on the interest and type factor for university[J].Journal of Zhejiang University of Technology,2019,47(4):425-429.

• [5] 张戈一，胡博然，常力恒，等.基于大数据分析挖掘的地质文献推荐方法研究[J].中国矿业，2017,26(9):92-97.ZHANG G Y,HU B R,CHANG L H,et al.Basics big date analysis analytic excavation geology reference recommendation method research[J].China Mining Magazine,2017,26(9):92-97.

• [6] 阮光册，谢凡，涂世文.基于Word2Vec的图书馆推荐系统多样性问题应用研究[J].图书馆杂志，2020,39(3):124-132.RUAN G C,XIE F,TU S W.Application research based on Word2Vec diversity in library recommender system[J].Library Journal,2020,39(3):124-132.

• [7] 耿立校，晋高杰，李亚函，等.基于改进内容过滤算法的高校图书馆文献资源个性化推荐研究[J].图书情报工作，2018,62(21):112-117.GENG L X,JIN G J,LI Y H,et al.Research on personalized recommendation of university library literature resources based on improved content-based filtering algorithm[J].Library and Information Service,2018,62(21):112-117.

• [8] 熊回香，孟璇，叶佳鑫.基于关键词语义类型和文献老化的学术论文推荐[J].现代情报，2021,41(1):13-23.XIONG H X,MENG X,YE J X.Recommendation of academic papers based on keyword semantic type and literature obsolescence[J].Journal of Modern Information,2021,41(1):13-23.

• [9] 陈长华，李小涛，邹小筑，等.融合Word2Vec与时间因素的馆藏学术论文推荐算法[J].图书馆论坛，2019,39(5):110-117.CHEN C H,LI X T,ZOU X Z,et al.A new Word2Vec algorithm of academic paper recommendation[J].Library Tribune,2019,39(5):110-117.

• [10] 杜永萍，杜晓燕，姚长青.基于主题效能的学术文献推荐算法[J].北京工业大学学报，2015,41(2):215-222.DU Y P,DU X Y,YAO C Q.Recommendation algorithm based on topic utility for academic papers[J].Journal of Beijing University of Technology,2015,41(2):215-222.[11]

• [11] 张卫卫，胡亚琦，翟广宇，等.基于LDA模型和Doc2vec的学术摘要聚类方法[J].计算机工程与应用，2020,56(6):180-185.ZHANG W W,HU Y Q,ZHAI G Y,et al.Academic abstract clustering method based on LDA model and Doc2vec[J].Computer Engineering and Applications,2020,56(6):180-185.

• [12] WANG Z Y,LIU Y,YANG J J,et al.A personalization-oriented academic literature recommendation method[J].Data Science Journal,2015,14:17.

• [13] KANAKIA A,SHEN Z H,EIDE D,et al.A scalable hybrid research paper recommender system for microsoft academic[C]//WWW'19:The World Wide Web Conference.New York:ACM,2019:2893-2899.

• [14] 黄泳航，汤庸，李春英，等.基于社区划分的学术论文推荐模型[J].计算机应用，2016,36(5):1279-1283.HUANG Y H,TANG Y,LI C Y,et al.Academic paper recommendation model based on community partition[J].Journal of Computer Applications,2016,36(5):1279-1283.

• [15] 贺超波，沈玉利，余建辉，等.基于学术社区的科技论文推荐方法[J].华南师范大学学报(自然科学版),2012,44(3):55-58.HE C B,SHEN Y L,YU J H,et al.Method for scientific paper recommendation based on academic community[J].Journal of South China Normal University(Natural Science Edition),2012,44(3):55-58.

• [16] 王庆，赵发珍.基于“用户画像”的图书馆资源推荐模式设计与分析[J].现代情报，2018,38(3):105-109.WANG Q,ZHAO F Z.Design and analysis of library resource recommendation model based on user profile[J].Journal of Modern Information,2018,38(3):105-109.

• [17] 黄立威，江碧涛，吕守业，等.基于深度学习的推荐系统研究综述[J].计算机学报，2018,41(7):1619-1647.HUANG L W,JIANG B T,LV S Y,et al.Survey on deep learning based recommender systems[J].Chinese Journal of Computers,2018,41(7):1619-1647.

• [18] 王妍，唐杰.基于深度学习的论文个性化推荐算法[J].中文信息学报，2018,32(4):114-119.WANG Y,TANG J.Deep learning-based personalized paper recommender system[J].Journal of Chinese Information Processing,2018,32(4):114-119.

• [19] 赵凯，王鸿源.LDA最优主题数选取方法研究：以CNKI文献为例[J].统计与决策，2020,36(16):175-179.ZHAO K,WANG H Y.Research on the selection method of LDA’s optimal topic number:taking CNKI literature as an example[J].Statistics & Decision,2020,36(16):175-179.

• [20] 邱均平.文献计量学[M].2版.北京：科学出版社，2019:67-70.

• [21] 刘伙玉.基于CNKI的图书、情报学与档案学学科文献半衰期分析[J].图书与情报，2015(1):106-111.LIU H Y.Analysis of half-life of literatures on libraries,information and archives based on CNKI[J].Library & Information,2015(1):106-111.

• [22] 何胜，熊太纯，柳益君，等.基于Spark的高校图书馆文献推荐方案及实证研究[J].图书情报工作，2017,61(23):129-137.HE S,XIONG T C,LIU Y J,et al.A spark-based scheme of university library literature recommendation and its empirical study[J].Library and Information Service,2017,61(23):129-137.